do
{
　　计算种群Pop中每一个体的适应度F(i)。
　　初始化空种群newPop
　　do
　　{
　　　　根据适应度以比例选择算法从种群Pop中选出2个个体
　　　　if ( random ( 0 , 1 ) < Pc )
　　　　{
　　　　　　对2个个体按交叉概率Pc执行交叉操作
　　　　}
　　　　if ( random ( 0 , 1 ) < Pm )
　　　　{
　　　　　　对2个个体按变异概率Pm执行变异操作
　　　　}
将2个新个体加入种群newPop中
} until ( M个子代被创建 )
用newPop取代Pop
}until ( 任何染色体得分超过Tf， 或繁殖代数超过G )

namespace GenticTSP
{
///<summary>
/// Fitness function for TSP task (Travaling Salasman Problem)
///</summary>
publicclass TSPFitnessFunction : IFitnessFunction
{
// map
privateint[,] map =null;

// Constructor
public TSPFitnessFunction(int[,] map)
{
this.map = map;
}

///<summary>
/// Evaluate chromosome – calculates its fitness value
///</summary>
publicdouble Evaluate(IChromosome chromosome)
{
return1/ (PathLength(chromosome) +1);
}

///<summary>
/// Translate genotype to phenotype
///</summary>
publicobject Translate(IChromosome chromosome)
{
return chromosome.ToString();
}

///<summary>
/// Calculate path length represented by the specified chromosome
///</summary>
publicdouble PathLength(IChromosome chromosome)
{
// salesman path
ushort[] path = ((PermutationChromosome)chromosome).Value;

// check path size
if (path.Length != map.GetLength(0))
{
thrownew ArgumentException(“Invalid path specified – not all cities are visited“);
}

// path length
int prev = path[0];
int curr = path[path.Length –1];

// calculate distance between the last and the first city
double dx = map[curr, 0] – map[prev, 0];
double dy = map[curr, 1] – map[prev, 1];
double pathLength = Math.Sqrt(dx * dx + dy * dy);

// calculate the path length from the first city to the last
for (int i =1, n = path.Length; i < n; i++)
{
// get current city
curr = path[i];

// calculate distance
dx = map[curr, 0] – map[prev, 0];
dy = map[curr, 1] – map[prev, 1];
pathLength += Math.Sqrt(dx * dx + dy * dy);

// put current city as previous
prev = curr;
}

return pathLength;
}
}
}

using AForge;
using AForge.Genetic;

namespace GenticTSP
{
class GenticTSP
{

staticvoid Main()
{
StreamReader reader =new StreamReader(“Data.txt“);

int citiesCount =31; //城市数

int[,] map =newint[citiesCount, 2];

for (int i =0; i < citiesCount; i++)
{
string value = reader.ReadLine();
string[] temp = value.Split(‘‘);
map[i, 0] =int.Parse(temp[0]); //读取城市坐标
map[i, 1] =int.Parse(temp[1]);
}

// create fitness function
TSPFitnessFunction fitnessFunction =new TSPFitnessFunction(map);

int populationSize = 1000; //种群最大规模

/*
* 0：EliteSelection算法
* 1：RankSelection算法
* 其他：RouletteWheelSelection 算法
* */
int selectionMethod =0;

// create population
Population population =new Population(populationSize,
new PermutationChromosome(citiesCount),
fitnessFunction,
(selectionMethod ==0) ? (ISelectionMethod)new EliteSelection() :
(selectionMethod ==1) ? (ISelectionMethod)new RankSelection() :
(ISelectionMethod)new RouletteWheelSelection()
);

// iterations
int iter =1;
int iterations =5000; //迭代最大周期

// loop
while (iter < iterations)
{
// run one epoch of genetic algorithm
population.RunEpoch();

// increase current iteration
iter++;
}

System.Console.WriteLine(“遍历路径是： {0}“, ((PermutationChromosome)population.BestChromosome).ToString());
System.Console.WriteLine(“总路程是：{0}“, fitnessFunction.PathLength(population.BestChromosome));
System.Console.Read();

}
}
}